I haven't been keeping any secrets. I've been testing search engine optimizations since January and have been verifying the results. Whenever I find anything of significance, I usually write about it.
I was very recently asked about my experiments with different types of comments and what I saw in the search engines. Some of my conclusions lead me to believe that many self-proclaimed SEO experts aren't experts at all.
Links Within Comments
Here's one example. We've been told that Google's algorithm can differentiate between links in a post and links in a comment. There are two occasions when that's true and one occasion when it's patently false. If the Googlebot follows a link that ends with something like "#comments", it knows the links are contained in comments. When the Googlebot goes to the second page of comments, for people who have comments set up on multiple pages, it knows. It doesn't know the difference, however, when it indexes the first page of the post, at least in the case of WordPress blogs. I can't tell you right now if other blogging platforms are affected in the same way.
Does it mean that the Google algorithm assigns less "weight" to links in comments? The answer to this is actually in the length of a comment, as well as the content of a comment. If a comment is of insufficient length, and the text between comment links are of insufficient length, a comment is guaranteed to carry less authority.
Comment Length
Comment length isn't a problem on most blogs. Only spam comments tend to be too short. Spam comments tend to be really short or really long.
I've been checking to see how much of a comment's content actually appear in the Google search engine results. I'm not sure how they do the math, but I see anywhere from 100 to 150 characters in the results.
Do you want to see this for yourself? Go to Google and search for "screaming traffic". It'll bring up one of my posts as the first result. Notice how many characters are in the description. Now, search for "screaming trade show displays". It'll bring up the same post as the first result. Look at how many characters are in the description. The difference in the descriptions is that the first result pulls the description from the post whereas the second result pulls the description from one of the comments. If comments aren't worth anything in the eyes of a search engine, how could this happen?
Allowing Comments
I don't know why people delete comments when they look a little suspicious. Some people delete comments when they don't contain a name and instead contain key words or the name of a website. It takes less than 30 seconds to click a link and check to make sure it doesn't point to a spam, porn, gambling or whatever site that you don't want to be associated with. I delete links like that with impunity.
If a comment doesn't fall within one of those categories, is either related to the topic of the post or another comment, it should be allowed to remain and I'll tell you why. It adds to your post and produces more keywords for the search engines to chew on. One comment can increase search visitors by as much as 10 times as the original post did. My infamous "Downloading Pirated Anything Is NOT Illegal" post continues to increase in search activity. I rarely delete any new comments that appear on it and I think I had a new comment yesterday. That post is approaching 15 months in age!
Comment Link Strength
Unless the original post ranks low in the search engines, it's doubtful that a comment link would have much effect on it. From my observations, it has very little effect on the site it links to unless the post itself is keyword optimized and is relative to the comments and the comment link. Does that make sense to you? It's confusing to me, so I'm sure it's confusing to you. "Screaming traffic" isn't related to "screaming trade show displays" except for the word screaming, which is way too popular of a word by itself to make a difference.
Backlinks from Comments
Comment links, when used on blogs that "dofollow", count as backlinks. According to Google, even if used on blogs that "nofollow", they can be followed and indexed by reference. I'm not sure how that works, but it happens. I've seen it. I saw one of my "nofollow" links within a post get followed and indexed. It may be because the post was linked to by another blog and that caused it, but I'll never know for sure.
All backlinks that are counted by the search engines carry "weight" or authority. The authority is determined by the linking relationship. In comments, it's actually easier to pick a related link than it is in a post. Since comment links can pointed to any page or post, I often wonder why people pick their home page more often than not. If your blog isn't related to mine, the link relationship between my post and your blog is not likely to be a strong one. However, if you pick a related post on your blog, the strength of that link more than doubles.
Any questions?
I don't know everything there is to know about how the search engines work and I suspect only the software engineers know every detail. Even if one so-called SEO expert knows what he or she's talking about, Google, Yahoo!, MSN/Live, AltaVista and other search engines could change their methods overnight and that information could no longer be true.
Is there anything that I missed? Is there a question I left unanswered? Feel free to throw tomatoes my way any time you like.



